Professional Documents
Culture Documents
20 The Lancet Volume 360 Issue 9328 2002 (Doi 10.1016/s0140-6736 (02) 09455-2) F Follath JGF Cleland H Just JGY Papp H Scholz K Peuhkurine - Efficacy and Safety of Intravenous Levosimendan Comp
20 The Lancet Volume 360 Issue 9328 2002 (Doi 10.1016/s0140-6736 (02) 09455-2) F Follath JGF Cleland H Just JGY Papp H Scholz K Peuhkurine - Efficacy and Safety of Intravenous Levosimendan Comp
Summary
Background Levosimendan, a novel calcium sensitiser,
improves myocardial contractility without causing an increase
in myocardial oxygen demand. We compared the effects of
levosimendan and dobutamine on haemodynamic performance
and clinical outcome in patients with low-output heart failure.
Methods Patients were recruited into a multicentre,
randomised, double-blind, double-dummy, parallel-group trial.
Under continuous haemodynamic monitoring, an initial loading
dose of levosimendan of 24 g/kg was infused over
10 min, followed by a continuous infusion of 01 g kg1 min1
for 24 h. Dobutamine was infused for 24 h at an initial dose of
5 g kg1 min1 without a loading dose. The infusion rate was
doubled if the response was inadequate at 2 h. The primary
endpoint was the proportion of patients with haemodynamic
improvement (defined as an increase of 30% or more in
cardiac output and a decrease of 25% or more in pulmonarycapillary wedge pressure) at 24 h. Analyses were by intention
to treat.
Findings 103 patients were assigned levosimendan and 100
dobutamine. The primary haemodynamic endpoint was
achieved in 29 (28%) levosimendan-group patients and 15
(15%) in the dobutamine group (hazard ratio 19 [95% CI
1133]; p=0022). At 180 days, 27 (26%) levosimendangroup patients had died, compared with 38 (38%) in the
dobutamine group (057 [034095]; p=0029).
Interpretation In patients with severe, low-output heart failure,
levosimendan improved haemodynamic performance more
effectively than dobutamine. This benefit was accompanied by
lower mortality in the levosimendan group than in the
dobutamine group for up to 180 days.
Lancet 2002; 360: 196202
196
Introduction
Heart failure is a common and growing problem associated
with frequent admissions to hospital and a poor prognosis.
It is the one most frequent reason for hospital admissions in
people aged over 65 years,1 and about 5000 hospital
admissions per million population per year are attributable
to or complicated by heart failure.2 Many of these hospital
stays are long; 5% or more of medical beds could be
occupied by patients with heart failure, accounting for more
than 50 000 hospital bed-days per million population per
year.2,3 About 40% of patients die within a year of an acute
exacerbation of severe heart failure.4
Management of an exacerbation of heart failure varies
according to its cause and presentation. Recognition and
treatment of precipitating factors such as rapid atrial
fibrillation, myocardial ischaemia, and infection are of key
importance. Acute cardiogenic pulmonary oedema is
frequently treated with intravenous diuretics and oxygen
alone, but opioids and nitrates probably have a valuable
role,5 although no substantial studies have been done.
When exacerbation of heart failure is accompanied by
hypotension, oliguria, or other evidence of a low cardiac
output, inotropic agents may be considered. However,
there is little evidence that intravenous inotropic therapy
with -agonists or phosphodiesterase inhibitors improves
outcome.6,7 A meta-analysis of trials of intravenous
inotropic agents suggested a non-significant trend towards
increased mortality,8 and studies of oral compounds have
consistently shown excess mortality.9,10
Levosimendan is a novel agent with a dual mechanism of
action developed for the treatment of decompensated heart
failure. This agent sensitises troponin C to calcium in a
manner dependent on calcium concentration, thereby
increasing the effects of calcium on cardiac myofilaments
during systole and improving contraction at low energy
cost.1115 Calcium concentration, and therefore sensitisation,
declines or is lost during diastole, allowing normal or
improved diastolic relaxation. Unlike agents that act
through adrenergic pathways, levosimendan does not cause
diastolic calcium overload, which can impair myocardial
relaxation, increase energy expenditure, or both,12,16 effects
that could result in the adverse effects of the latter classes of
inotropic agents. Levosimendan also leads to vasodilatation
through the opening of ATP-sensitive potassium
channels.17 By these inotropic and vasodilatory actions,
levosimendan increases cardiac output without increasing
myocardial oxygen demand.12,13,1719
Dose-ranging studies, with infusion times from 6 h to
24 h, have already shown that levosimendan improves
cardiac haemodynamics and offers symptomatic benefit in
patients with severe heart failure.20,21 Although placebocontrolled studies of dobutamine have shown that this drug
improves haemodynamic variables, outcome data are
few.8,22 We compared the effects of dobutamine and
levosimendan on haemodynamics over 24 h and then
prospectively followed up clinical outcome over the next
31 days. Furthermore, at the request of the regulatory
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
Methods
Patients
This multicentre, randomised, double-blind, doubledummy, parallel-group trial was done at 26 centres in
11 European countries (Austria, Denmark, Finland,
France, Germany, Hungary, Italy, Switzerland, the
Netherlands, Sweden, and the UK). The primary objective
was to compare the proportions of patients with haemodynamic improvement at the end of the 24-h infusion
period in the groups assigned dobutamine or levosimendan.
Eligible patients were those admitted to hospital with
low-output heart failure who were judged to require
haemodynamic monitoring and treatment with an
intravenous inotropic agent. Patients in one or more of the
following clinical categories could be included:
deterioration of severe chronic heart failure despite
optimum oral therapy with vasodilators and diuretics,
including those awaiting cardiac transplantation; severe
heart failure after cardiac surgery; or acute heart failure
related to a cardiac or non-cardiac disorder of recent onset.
In addition, patients had to have a left-ventricular ejection
fraction of less than 035 as shown by two-dimensional
echocardiography or radionuclide ventriculography within
1 month of study enrolment, an actual cardiac index of less
than 25 L min1 m2, and a mean pulmonary-capillary
wedge pressure of more than 15 mm Hg.
Exclusion criteria were: age younger than 21 years; childbearing potential; heart failure due to restrictive or
hypertrophic cardiomyopathy or to uncorrected stenotic
valvular disease; chest pain at the time of randomisation;
sustained ventricular tachycardia or ventricular fibrillation
within the previous 2 weeks; atrioventricular block of
second or third degree; heart rate more than 120 beats/min
at rest; systolic blood pressure below 85 mm Hg; severe
renal failure (serum creatinine >450 mol/L); hepatic
failure; cardiac tamponade; adult respiratory distress
syndrome; and septic shock. The protocol prohibited
intravenous -adrenergic agonists within 30 min before
baseline haemodynamic measurements, intravenous
vasodilators within 2 h, intravenous milrinone or
enoximone within 12 h, and intravenous amrinone within
2 days. The timing of other cardiovascular drugs (such
as digoxin, diuretics, angiotensin-converting-enzyme
inhibitors, and other vasodilators) was standardised to
minimise any effect on haemodynamic measurements.
These drugs had to be given at least 6 h before baseline
measurements, between 4 h and 18 h of the study period,
or after the end of the study drug infusion. In general, the
dose of these concomitant medications was held constant,
unless urgent modifications were required on clinical or
haemodynamic grounds. Written informed consent was
obtained at the time of initial screening before any
haemodynamic assessments.
Protocol
To maintain masking, each patient received two
simultaneous infusions, one active and one placebo. The
placebo infusion regimen was identical to its respective
active counterpart. Patients were enrolled by the study
investigators and assigned randomly to the levosimendan
group or the dobutamine group in blocks of four according
to a computer-generated code created by Orion Pharma for
each centre. Treatment allocation and size of the randomisation blocks were concealed from the investigators.
The recruiting investigator received a sealed envelope for
197
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
Dobutamine
(n=100)
103 assigned
levosimendan
Levosimendan
(n=103)
Demography
Mean (SD) age, years
M/F
60 (11)
85/15
58 (11)
91/12
50 (50%)
50 (50%)
46 (45%)
57 (55%)
Clinical presentation
Deterioration of chronic heart failure*
Acute heart failure
Postoperative exacerbation
Awaiting heart transplant*
92 (92%)
0
2 (2%)
15 (15%)
86 (84%)
7 (7%)
4 (4%)
19 (18%)
203 randomised
100 assigned
dobutamine
4 (17)
366 (104)
191 (044)
48 (18)
24 (7)
55 (17)
373 (078)
194 (036)
47 (13)
25 (8)
98 (64)
117 (19)
71 (12)
25 (8)
104 (67)
112 (18)
69 (12)
24 (6)
81 (16)
82 (15)
75 (75%)
90 (90%)
88 (88%)
39 (39%)
50 (50%)
42 (42%)
13 (13%)
4 (4%)
2 (2%)
78 (76%)
98 (95%)
92 (89%)
38 (37%)
33 (32%)
45 (44%)
16 (16%)
4 (4%)
0
198
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
Median change
p*
20
Variable
Cardiac output, L/min
080
Pulmonary-wedge pressure (mm Hg) 3
Stroke volume, mL
8
Pulmonary-artery diastolic pressure 3
(mm Hg)
Systemic vascular resistance
46
(mmHg L1 min1)
109
7
10
6
0048
0003
0262
0001
58
0068
Dobutamine Levosimendan
15
10
05
0
Pulmonary-capillary wedge
pressure (mm Hg)
Dobutamine
Levosimendan
h
30
h
24
h
4
h
Time
Figure 3: Change in cardiac output and pulmonary-capillary
wedge pressure from baseline to 30 h
With -blockers
p=001
10
05
n=69
n=33
Levosimendan
n=67
n=28
Dobutamine
0
Change in pulmonary-capillary
pressure (mm Hg)
23
.5
Without -blockers
15
10
m
i
1n
h
h
p=003
Results
Characteristics of study groups
203 patients were enrolled between Jan 2, 1997, and
Nov 3, 1998: 103 were assigned treatment with levosimendan and 100 dobutamine (figure 1). Baseline
characteristics of the two groups are summarised in
table 1. Four patients (one assigned levosimendan, three
dobutamine) did not receive an infusion of the study drug
(catheter insertion failed or consent was withdrawn during
catheter insertion) and did not undergo any haemodynamic
measurements. These patients were included in all efficacy
and safety analyses in accordance with the intention-totreat principle and were classified as non-responders for the
analysis of the primary endpoint. They were also included
in the haemodynamic analyses and assigned worst rank,
and for the analyses of death and worsening heart failure
they were included as if they had received the study drug.
Five patients in the levosimendan group and six in
the dobutamine group had dose-limiting events leading
to temporary discontinuation of study medications
(figure 1). 16 patients did not receive study drugs for the
planned duration of treatment (ten dobutamine, six
levosimendan).
199
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
100
Levosimendan
90
80
70
60
Dobutamine
50
40
30
20
90
10
0
11
0
12
0
13
0
14
0
15
0
16
0
17
0
18
0
70
80
50
60
30
40
0
10
20
10
0
Time (days)
Numbers at risk
100 94 91 85 82 81 81 80 78 77 75 74 74 72 67 64 63 62 62
Dobutamine
Levosimendan 103 101 97 96 94 92 91 90 90 90 90 88 88 87 87 83 80 77 76
Haemodynamic effects
A significantly greater proportion of patients in the
levosimendan group than in the dobutamine group
achieved the primary endpoint; at 24 h, haemodynamic
performance had improved in 29 (28%) versus 15 (15%)
patients (hazard ratio 19 [95% CI 1133]; p=0022).
We did an additional per-protocol analysis on all
randomised patients who received their study drug for at
least 18 h and had at least two baseline and 24 h
haemodynamic measurements (n=94 for levosimendan,
n=88 for dobutamine). Patients who received no study
drug, those who did not receive the drug for at least 18 h
(figure 1), and those for whom baseline measurements were
missing (two levosimendan, one dobutamine) were
excluded from this analysis (nine levosimendan, 12
dobutamine). One patient who received a 226 h infusion of
dobutamine was included in the per-protocol analysis but
was also classified as a permanent withdrawal; another
patient who received a 24 h infusion of dobutamine but
withdrew consent at this time was also included in the perprotocol analysis. 29 (31%) of 94 patients in the
levosimendan group and 13 (15%) of 88 in the dobutamine
group achieved the primary endpoint at 24 h (hazard ratio
209 [116375]; p=0021).
Levosimendan treatment had a consistently better effect
than dobutamine on the individual haemodynamic
variables at the end of the 24 h treatment period (table 2).
Both levosimendan and dobutamine increased heart rate by
a modest and similar amount (57 [SD 121] vs 40 [98]
beats/min). Serum creatinine concentration and markers of
liver dysfunction declined in the levosimendan group
compared with the dobutamine group, possibly reflecting a
superior effect on organ perfusion.
Clinical symptoms of dyspnoea and fatigue tended to
improve to a greater extent with levosimendan than with
dobutamine, but these differences were not significant.
Dyspnoea improved in 52 (68%) of 76 and 44 (59%) of 74
patients with baseline symptoms in the levosimendan and
dobutamine groups, respectively (p=0865); fatigue
improved in 50 (63%) of 79 and 37 (47%) of 79
(p=0155).
The use of -blockers had significant effects on the
increase in cardiac output and the decrease in pulmonarycapillary wedge pressure (p=001 and p=003, respectively;
figure 2). -blockade attenuated the effect of dobutamine
on cardiac output and pulmonary-wedge pressure, but did
200
p*
Dobutamine
(n=100)
Levosimendan
(n=103)
42
48
088
Disorder aggravated
044
Headache or migraine
14
0052
0013
4
1
3
9
0
1
0252
NA
0207
13
1
3
2
3
2
1
1
4
2
1
0
0
1
1
1
0023
10
0364
NA
NA
0618
10
10
Gastrointestinal disorders
0098
5
3
10
1
0
6
NA
NA
NA
NA=not analysed. Some patients had more than one adverse event. *Fisher's
exact test.
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
Change in variable
Dobutamine
Levosimendan
Dobutamine
Levosimendan
Haematology*
Haemoglobin (g/dL)
Erythrocytes (1012/L)
Leucocytes (109/L)
Platelets (109/L)
91
80
91
91
96
87
96
96
04 (09)
013 (033)
051 (183)
218 (343)
08 (11)
027 (042)
014 (234)
265 (293)
0005
005
011
046
Clinical chemistry*
Serum sodium (mmol/L)
Serum potassium (mmol/L)
Fasting blood glucose (mmol/L)
86
85
78
93
95
83
1 (4)
02 (05)
04 (34)
1 (4)
01 (08)
01 (25)
089
005
023
88
77
86
87
78
96
84
91
92
84
1 (-12 to 9)
3 (-14 to 7)
2 (-6 to 1)
2 (-6 to 1)
3 (-15 to 1)
9 (19 to 2)
8 (21 to 3)
4 (9 to -1)
25 (-8 to -1)
6 (13 to 0)
003
001
006
003
022
*Values are mean (SD) and p values are calculated by ANCOVA. Values are median (IQR) and p values are calculated by Mann-Whitney sum of ranks test.
Discussion
In this study, there was greater haemodynamic improvement in patients with severe low-output heart failure with a
24 h infusion of levosimendan than with dobutamine. At
the doses studied, levosimendan was associated with a
slightly greater increase in cardiac output, and more
pronounced decreases in pulmonary-capillary wedge
pressure and systemic vascular resistance than dobutamine.
These findings are consistent with those of other studies20
and with the known positive inotropic and peripheral
vasodilatory effects of levosimendan. Levosimendan was
well tolerated; it caused fewer cardiac adverse events than
dobutamine.
The haemodynamic effects of levosimendan, unlike those
of dobutamine, were not attenuated by the concomitant use
of -blockers. This finding is important, in view of the
increasing evidence for and the usefulness of -blockers for
the management of severe heart failure. This study suggests
that levosimendan could be a useful agent for the management of exacerbations of heart failure during uptitration or
long-term maintenance therapy with -blockers. There was
little difference in the reported improvement of symptoms
201
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
10
11
12
13
14
Contributors
15
16
17
Study investigators
DenmarkP Thayssen.
FinlandK Peuhkurinen, M Niemel, A Ylitalo, J Laurila, O Takkunen,
V P Harjola, V Y O Pettil, J Hartikainen, O Parviainen, E Ruokonen,
H Tulla, R Suojaranta-Ylinen, J Takala, K Korpilahti, M Lindroos,
R Jussila.
FranceP Gibelin, J-L Dubois-Rande, S Benacerraf, I Macquin-Mavier.
GermanyR Zotz, R Pliquett, H Just, B Sauerbier, H-P Hermann,
N Bruns, H Baberg, V Mitrovic, M Hamel, W Sehnert, M Keck,
K Neumann, E Altmann, A Graf, F Hartmann.
HungaryI des, K Csap, A Bdi, M Csandy, M Hogye, J Szecsi,
C Lszl, P Katalin, G Argyelan, L Cserhalmi, L Vandor, T Sznyi,
O Partos,
A Rnaszki, K Karlczai, M Somloi.
NetherlandsA Balk, F Jonkman, C van der Zwaan, P Dunselman.
SwedenC-H Berg, K-E Karlberg, I Hagerman, P Kvidal, G Wikstrm.
UKJ G F Cleland, J McGowan, K Sherry, J Ross.
18
19
20
21
22
23
Acknowledgments
We thank the study investigators for their contribution to the study and care
of patients at their individual institutions, Toni Sarapohja (Orion Pharma,
Espoo, Finland) for supervising data management and for providing
assistance in statistical issues, Evelyn Wesby-van Swaay (Orion Pharma,
Espoo, Finland) for assistance in medical writing of the clinical study report,
and Rajni Dogra (London, UK) for writing and editorial support.
24
25
References
1
2
3
4
202
26
27
28
For personal use. Only reproduce with permission from The Lancet Publishing Group.